It is 11 p.m. on a Sunday and the Monday morning client report is pulling in zeros. Not wrong numbers, zeros. The IMPORTRANGE formula that has quietly fed the marketing dashboard for eight months stopped working sometime between Friday and now, and there is no error message, no alert, nothing but a blank column where last week's ad spend used to be.
This is the moment every agency ops person eventually has with a Google Sheets reporting automation setup. It is not a dramatic failure. It is a small, silent one, and it always happens right before someone needs the number.
TL;DR:
- A standard GA4 property is capped at 200,000 core tokens per day and 40,000 per hour, a limit that a Sheets add-on refreshing several reports on a short schedule can hit without anyone changing anything on purpose.
- Apps Script time-based triggers get 90 minutes of total daily runtime on a free consumer Google account, versus 6 hours a day on Google Workspace, which is why the exact same automation script behaves differently depending on which kind of account runs it.
- Paid Sheets-connector tools add their own subscription on top of "free" Sheets. Coefficient's own pricing page lists a Starter plan at $49 a month and a Pro plan at $99 per user per month once a team outgrows the free tier's refresh limits.
- As an illustration, not a universal number: a person maintaining five client reports, each pulling from two or three sources with roughly ten minutes of manual fixing per cycle, is already spending close to an hour a week on upkeep before adding anything new.
What "Google Sheets reporting automation" actually covers
Three different setups all get called this, and they fail in different ways.
The first is formula-based pulling: IMPORTRANGE to stitch sheets together, IMPORTDATA or IMPORTXML to grab something from a URL, or a native connector like Google's own Analytics add-on. No code, no script, just functions sitting in cells.
The second is Apps Script: a script attached to the spreadsheet that runs on a timer, calls an API, and writes the results into cells. This is the step up most agencies take once formulas alone cannot reach a data source.
The third is a paid Sheets-sync tool, something like Coefficient or Supermetrics, that handles the API connection and refresh scheduling for you in exchange for a subscription.
Each one trades a different kind of cost for a different kind of convenience, and almost nobody picks between them deliberately. Most reporting setups start with formulas, grow an Apps Script patch when a formula cannot reach a source, and only add a paid connector once the Apps Script patch becomes someone's unofficial part-time job.
Where the free, built-in path breaks: Apps Script's own limits
A time-based Apps Script trigger is the thing most "automate your Google Sheets reports" tutorials tell you to set up, and it works well right up until the daily runtime quota.
On a free, personal Google account, Apps Script trigger runtime is capped at 90 minutes of total execution time per day. A Google Workspace account gets 6 hours a day instead, 4x more headroom, which most people setting up their first automation have no reason to know exists. Individual executions are also capped at 6 minutes each, so a script pulling a large dataset has to be written to finish inside that window or split the work across multiple runs.
None of this shows up as an error on day one. It shows up weeks or months later, once more scripts get added to the same account and the daily total quietly creeps past the limit, at which point some subset of the day's scheduled refreshes simply do not run, with no notification sent to anyone.
Where the API path breaks: quota limits that scale with how many reports you run
Pulling GA4 data into a sheet, whether through Google's own connector, a script, or a third-party add-on, draws from the same quota pool regardless of which tool makes the request. A standard property is limited to 200,000 core tokens per day, 40,000 per hour, and 14,000 per project per hour.
That sounds like a lot until several reports are refreshing on the same schedule. Five client reports each refreshing hourly, each making a handful of requests per refresh, is a different load than one report refreshing once a day, and the quota does not scale with how many clients you have added since the setup was first built.
The cost nobody adds to the "free" spreadsheet
The point where most agencies reach for a paid connector is the point where formulas and Apps Script both stop being worth the maintenance. Coefficient's pricing page prices a Starter plan at $49 a month for "solo builders ready to operationalize their manual workflows," and a Pro plan at $99 per user per month once a team needs more automation and larger data pulls. Supermetrics and similar tools follow the same shape: free or cheap at small scale, a real line item once refresh frequency or data volume grows.
That subscription is not the full cost either. Someone still has to notice when a pull silently stops working, someone still has to onboard every new client's data source into the same sheet structure, and someone still has to explain to a client why last Tuesday's number looked different from what shipped in the final report. None of that shows up on the connector's pricing page, and all of it is the real reason spreadsheet-based reporting gets expensive at the exact moment it looks "free."
The actual threshold, not a number but a question
There is no fixed client count or data-source count where Google Sheets definitively stops working, and any guide claiming otherwise is making it up. The question that actually predicts the breaking point is simpler: what happens the one time a pull fails and nobody notices for three days.
If the answer is "a client's monthly report is late and someone apologizes," the spreadsheet is probably still fine. If the answer is "a decision gets made on a number that was already wrong when it was pulled," the cost of staying on a formula-and-script setup has quietly become bigger than the cost of building something that is designed not to fail silently in the first place, the kind of system our data analytics team builds when a client outgrows the spreadsheet.
A related decision shows up on the automation side, not just the reporting side. The same Apps Script trigger limits that cap a report refresh also cap any other scheduled task layered onto a spreadsheet, like a script that moves leads between tabs or fires a notification. Once that kind of logic needs to run reliably across dozens of operations a day rather than a handful of report refreshes, it belongs in a workflow automation platform built to run jobs on a schedule without a daily runtime ceiling, not bolted onto a spreadsheet that was never meant to be a task runner.
Key takeaways
- Formula pulls, Apps Script, and paid connector tools are three different automation layers with three different breaking points, not one technique with one limit.
- GA4's own quota, 200,000 tokens a day and 40,000 an hour on a standard property, applies no matter which tool is making the request on your behalf.
- Apps Script trigger runtime is capped at 90 minutes a day on a free account versus 6 hours a day on Google Workspace, a gap that explains why the same script can behave completely differently for two different users.
- Paid Sheets connectors start around $49 to $99 a month once free-tier limits are hit, and that subscription is on top of, not instead of, the time someone still spends catching and fixing silent failures.
- The real signal to move off a spreadsheet is not a client count. It is whether a silent failure would just be late, or would quietly feed a wrong decision before anyone caught it.
